This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

IWR1443BOOST: AR_SCL, AR_SDA pull resistor and decoupling CAP?

Part Number: IWR1443BOOST
Other Parts Discussed in Thread: IWR1443,

Anyone know why the pull up resistors for AR_SDA and AR_SCL change from 10k ohms in rev_A schematic to 2.94k ohms in rev_B schematic? This change increase the current draw of Rev B board; I also noticed the decoupling capacitors are reduced for LDO 1.8V and 1.3V output in rev_B schematic, any good reason? (We like to reduce those decoupling CAP, but we need to know if it could degrade the performance)

Thanks

  • Hello Feng,

    10k pull-up will still work, but may not support fast mode depending on the load capacitance of your bus. 

    The output caps on the LDO were not actually reduced, they were just moved to the LC filter. If you look at the LC filter you will see that there is a 1uF and 10uF cap on each filter. These are still connected to the output net of the LDOs. If you plan to use the LDOs you should not reduce the number of output caps.

    Regards,

    Adrian

  • Hi Adrian,

    Thanks for the information! Do you ever test the fast mode with 10k ohms resistor you? I also noticed one ferrite bead is added to the schematic in rev B between  PMIC1.2V and AR_1.2V, how much improvement by adding this ferrite bead? We try to minimize using the inductance and the capacitance in our design.

    Thanks,

  • Hello Feng,

    The ferrite bead was added in place of the LDO so that we can filter out the 4MHz switching spurs from the PMIC. If you use the LDOs then you can omit the ferrite beads. 

    We have not tested fast mode with the 10k resistor, but you can approximate the maximum resistor value that your system can support using equation 6 in the below app note using a rise time of 300ns for fast mode:

    Regards,

    Adrian

  • Hi Adrian,

    Thanks for the information! I use the equation from you attached file, the max pull resistor for standard mode is around 2.9K ohms, the pull up resistor for fast mode is 0.88k ohms, neither mode can use 10k ohms. In rev A, the pull up resistor is 10k ohms and in modified Rev A with ES3 chip (pre rev B board), the pull up resistor is also 10k ohms, they all work fine. When this equation come out? Do you see any problem with rev B board by using 10 k ohms resistor?

    Thanks and best regards,

  • Hello Feng,

    0.88k is around 400pF bus capacitance. You should be expecting well below 400pF capacitance unless you have a significant amount of devices on the bus and very long traces. ~10k pullups for slow and 2-3k pullups for fast shouldn't be an issue in most cases.

    Regards,

    Adrian

  • Hi Adrian,

    The IWR1443 spec says "line capacitance 400pF, but it does not give the value for Tr(rising time) on I2C data sheet, 10us clock for slow, 2.5us for fast. the length of the traces is same as IWR1443boost, but we use two 20 pin connectors on IWR1443boost, we try to use I2C to transfer the data to MSP board (control board), but we need meet the low current requirement (4mA to 20mA). Where you get the value for Tr? All TI's I2C Tr time is same as the PDF file you attached, like 1000ns for low, 300ns for fast? Do you think I2C is good choice to transfer data to MSP control?

    Thanks and best regards,

  • Hello Feng,

    400pF is the maximum bus capacitance. This will not be the capacitance on a typical bus, you will need to estimate this based on your PCB but 10k and 2k pullups are a general rule of thumb. 1000ns and 300ns rise times are specs defined by the I2C standard. You will need to determine what data you are trying to send and whether or not the I2C interface is suitable for this. 

    Regards,

    Adrian

  • Thanks you very much!